=== Hussain is now known as Guest62667 === daker_ is now known as daker [10:47] <_mup_> Bug #830995 was filed: default-instance-type and default-image-id are unhelpful < https://launchpad.net/bugs/830995 > [10:52] <_mup_> Bug #830999 was filed: cannot specify machine constraints < https://launchpad.net/bugs/830999 > [13:18] I need to learn to pay attention to my status :/ [13:20] <_mup_> Bug #831058 was filed: cobbler FileStorage assumes no authentication required < https://launchpad.net/bugs/831058 > [13:38] Hello all [13:38] Good week kickoffs for everybody [13:49] niemeyer: heyhey! [13:49] niemeyer: how's it going? [13:51] fwereade: Hey! [13:51] fwereade: Very nice [13:51] fwereade: Had a good relaxing weekend [13:51] fwereade: You? [13:51] niemeyer: pretty good thanks [13:51] niemeyer: keeping half an eye on libya, what with it being right in my back yard [13:52] fwereade: :( [13:53] niemeyer: it seems to me like a tentative ":)" at the moment, but who knows what's coming :/ [13:54] niemeyer: and we didn't give back his fighters, so I doubt we have much to fear from the new lot :p [13:54] niemeyer: anyway :) [13:54] niemeyer: I've been wondering about bug importance and milestones [13:54] fwereade: Yeah.. worth paying some attention to [13:55] fwereade: Ok [13:55] niemeyer: there are a few bugs marked "high" but not assigned to eureka -- is that explicit notice that we shouldn't be worrying about them now? [13:58] fwereade: It's a good hint they might either a) be misplaced; or b) be miscategorized [13:58] niemeyer: hm, ok :) [13:58] niemeyer: I've been a little reluctant to go around assigning priorities and suchlike, what with still being pretty new [13:59] niemeyer: which is why 90% of my branches have been fixes to "uncategorized" bugs [13:59] niemeyer: should I perhaps be taking a different approach? [14:00] fwereade: I would guess this happens due to our overuse of bugs to mean "task" [14:00] fwereade: You're working on a single High priority problem [14:00] fwereade: Getting physical deployments working [14:00] niemeyer: similarly, when I report a bug, I don't usually assign a milestone; I only do that when I'm actually working on something, but I'm suddenly worrying that they'll just fly under the radar [14:01] fwereade: You can tons of uncategorized bugs because they are simply a path to get pieces of that work landed [14:01] s/You can/You get/ [14:01] niemeyer: makes sense [14:01] fwereade: The bug priority is most interesting when it's an actual issue we want to stash away without an assignee [14:01] fwereade: Because it tells the next guy what he should be paying attention to [14:03] niemeyer: I probably kinda mostly see [14:03] niemeyer: as an example, there's this idea that we should be able to specify more about what sort of machines are used in certain circumstances [14:04] niemeyer: a bug for that didn't leap out at me, so I filed one this morning, as a general "we should do this sometime" note [14:04] fwereade: That sentence gives such a positive feeling :-) [14:04] niemeyer: haha :) [14:05] that's lp:830999, which has no importance, assignee, or milestone [14:06] niemeyer: should it maybe be High, None, Eureka? [14:06] fwereade: Definitely not Eureka [14:06] fwereade: I'd personally qualify that as a Medium, non-Eureka [14:06] fwereade: Eureka has a short time span (as usual ;-).. [14:06] niemeyer: I feel vindicated in my reluctance to go around categorizing things myself ;) [14:07] niemeyer: I don't think I know when eureka actually is targeted to... LP thinks it was 1/1/2011, which I suspect to be incorrect (:p) [14:07] fwereade: As far as physical deployment goes, Eureka is about finishing the barebones [14:07] fwereade: Making it work reliably [14:08] fwereade: Rather than getting fancy features in [14:08] fwereade: Eureka is 11.10 [14:08] niemeyer: that's fine by me ;) [14:08] fwereade: End of September, pretty much [14:09] niemeyer: ok, makes sense, and we have some sort of special dispensation to release whatever we have done (while most things have already passed the deadline to get in) ...right? [14:10] fwereade: Kind of.. this is actually the real deadline [14:10] fwereade: For everyone [14:10] fwereade: Or rather.. for everyone in similar circumstances [14:11] fwereade: E.g. Ensemble is in universe, it's already there right now, etc [14:20] i think the close out of the milestone is scheduled right before the final testing freeze [14:20] hmm.. actually its not, i should move it up [14:21] the final freeze is sept 22nd [14:21] er 29th [14:21] 22nd is beta 2 [14:37] hazmat: SpamapS requested us to land things by Sep 22nd for testing.. [14:37] hazmat: We can close the milestone with 29th [14:38] hazmat: and have that week for testing on our side too [14:38] Still within Eureka [15:04] negronjl: ok, I tried a few times over the weekend and just now, still getting this "error: { "$err" : "not master", "code" : 10107 }" every time [15:28] Lunch time.. [15:34] Ensemble is on devops cafe! http://devopscafe.org/show/2011/8/22/devops-drop-004.html [15:36] nice [15:40] jcastro: did they talk about it? [15:40] yeah [15:40] it's a short podcast, worth a listen [15:43] jcastro: interesting..thanks [16:02] <_mup_> ensemble/expose-cleanup r328 committed by jim.baker@canonical.com [16:02] <_mup_> Initial refactoring per review [16:03] <_mup_> ensemble/expose-cleanup r329 committed by jim.baker@canonical.com [16:03] <_mup_> Merged trunk [16:07] jcastro: I'll look into it. [16:07] negronjl: ok I tried some other things and that didn't work either [16:08] I have the instance up and running now if you want to check it out [16:08] sure. add my ssh-id ( negronjl ) [16:24] lol... John Willis [16:24] love his monitoring lightning talk [16:27] jcastro: How is it in the podcast? [16:27] starts around 4:50 in [16:28] * niemeyer downloads [16:28] Mostly curious about what the take was [16:28] and what's the "ironic twist" :) [16:28] Not sure [16:28] still listening [16:30] I have like a minute of silence in mine [16:31] yeah [16:31] thats a quote [16:31] turn it up [16:31] its from Fast Times @ Ridgemont High [16:31] whoa I got a silence at 8:49 too [16:32] right at the juicy part ;) [16:32] its like the missing fuzz on the nixon tapes. ;) [16:34] yeah that missing minute is really weird [16:40] Nice.. quite positive overall [16:43] clearly he'll have to do another one with more detail. :) [16:43] Maybe one of his listeners will write a formula w/ Chef. ;) [17:10] ensemble team, wikimedia, check emails [17:10] happening now [17:38] <_mup_> ensemble/expose-cleanup r330 committed by jim.baker@canonical.com [17:38] <_mup_> Added mock tests with respect to refactoring of remove_security_groups [17:54] fwereade: ping... [17:54] fwereade: has something change whithin the code that generates the cloud-init user data? [18:04] w/in 14 [18:11] * SpamapS cheers for adam_g's w/inning [18:14] <_mup_> ensemble/expose-cleanup r331 committed by jim.baker@canonical.com [18:14] <_mup_> PEP8, PyFlakes, call remove_security_groups more robustly [18:44] so dumb question: have we tried testing ensemble against Oneiric images in EC2/OpenStack...or are we still using Natty-based amis [18:54] robbiew, we've tried it, its not the default [18:55] bug 816169 has broken a few peoples' formulas.. [18:55] <_mup_> Bug #816169: When using Ensemble, add-apt-repository no longer functions properly

< https://launchpad.net/bugs/816169 > [18:55] need to tak a look at that pronto [18:56] we should probably switch that to oneiric, there's a bug open about using the default package distros instead of the ppa [18:56] SpamapS, i thought i merged the fix for that [18:56] oh.. nevermind [18:56] might be good to switch it once beta is out [18:57] SpamapS, there was another bug about pythonpath env variable leaking to executed hooks, and causing issues with package installation. that's no longer the case though [19:00] SpamapS: Do you know why they broke? [19:00] SpamapS: There's a thread with some questioning around that [19:01] adam_g mentioned the interactivity, but Michael Vogt mentioned it actually checks for the tty presence [19:01] Not sure about what's happening there [19:02] niemeyer: its actually cloud-init that was screwing with the tty check [19:02] I have not looked at it, just followed the thread. [19:03] SpamapS: just fixed that cloud-init bug, will check on the other python problems soon [19:04] niemeyer: the interactivity change is unrelated to #816169 that SpamapS mentioned [19:04] <_mup_> Bug #816169: When using Ensemble, add-apt-repository no longer functions properly

< https://launchpad.net/bugs/816169 > [19:16] SpamapS: looks like that is no longer an issue. seems there was some weirndess in the python environment that has since settled? im going to wait for cloud-init to be updated in images so i can bootstrap+deploy without hacking before confirming 100% [19:17] adam_g: so since we don't really know where the problem was.. maybe just close all the tasks as Invalid once you've confirmed it doesn't exist anymore? [19:17] SpamapS: will do [19:19] adam_g: sweet thx! [19:26] * robbiew needs to eat...back in a bit [19:36] fwereade: when you are around pelase give me a shout [19:51] adam_g, yeah.. it was probably fixed as part of bug 816264 [19:51] <_mup_> Bug #816264: PYTHONPATH causing corrupt environment < https://launchpad.net/bugs/816264 > [19:58] hazmat: ah, nice [20:31] I'll break for some coffee [20:31] biab [20:34] niemeyer: your Google GOpher just arrived at my house [21:02] robbiew: Hah :) [21:02] niemeyer: I'll be sure to bring it to UDS [21:02] squishable.com....never heard of them before [21:03] but apparently they are pretty popular [21:03] the GOpher must be "special" [21:03] lol [21:04] robbiew: They are, somehow.. :-) [21:05] robbiew: Thanks for feeding it, meanwhile ;) [21:05] lol...sure thing [21:10] I thought they were going to send the gopher here as they asked for my home address, but if they actually sent the gopher to the original address I provided, I wonder what they've put in the mail now.. [21:40] Today was remarkably unproductive === otubo is now known as otubo[AFK] [22:43] <_mup_> ensemble/lxc-lib-merge r323 committed by kapil.thangavelu@canonical.com [22:43] <_mup_> merge lxc-lib [22:45] <_mup_> ensemble/formula-state-with-url r315 committed by kapil.thangavelu@canonical.com [22:45] <_mup_> parameterize trunk in ENSEMBLE_TRUNK env variable for makefile [22:45] <_mup_> ensemble/formula-state-with-url r316 committed by kapil.thangavelu@canonical.com [22:45] <_mup_> link reference doc for s3 auth url === m_3 is now known as mmm === mmm is now known as m_3 [23:02] <_mup_> ensemble/formula-state-with-url r317 committed by kapil.thangavelu@canonical.com [23:02] <_mup_> formula urls are required, also seed ec2 test provider base class with fake credentials [23:03] <_mup_> ensemble/formula-state-with-url r318 committed by kapil.thangavelu@canonical.com [23:03] <_mup_> merge trunk [23:06] <_mup_> ensemble/trunk r323 committed by kapil.thangavelu@canonical.com [23:06] <_mup_> merge formula-state-with-url [r=niemeyer,bcsaller][f=761053] [23:06] <_mup_> Formula states are now created in an enviroment with a url to their [23:06] <_mup_> location in provider storage. This is to enable machine agents not to [23:06] <_mup_> need access to the provider for deploying formulas. [23:14] <_mup_> ensemble/machine-agent-uses-formula-url r316 committed by kapil.thangavelu@canonical.com [23:14] <_mup_> merge trunk [23:21] <_mup_> ensemble/trunk r324 committed by kapil.thangavelu@canonical.com [23:21] <_mup_> merge machine-agent-uses-formula-url [r=niemeyer,fwereade][f=828189] [23:21] <_mup_> Machine agents download formulas using the formula url, instead of [23:21] <_mup_> retrieving them directly from provider storage, when deploying units. [23:27] <_mup_> Bug #831688 was filed: Openstack compatibility is needed < https://launchpad.net/bugs/831688 > [23:31] Can someone please have a second review on this: https://code.launchpad.net/~fwereade/ensemble/cobbler-zk-connect-error-messages/+merge/72033? [23:32] <_mup_> ensemble/stack-crack r326 committed by kapil.thangavelu@canonical.com [23:32] <_mup_> formulas are published with a formulas- prefix instead of a formulas/, per workaround for bug 829880 [23:44] <_mup_> ensemble/flex-origin r323 committed by jim.baker@canonical.com [23:44] <_mup_> Fix test that was depending on hash ordering in dict