[07:17] Hi === otubo is now known as otubo[AFK] [09:45] Nice, we're on http://www.linuxtoday.com/ === ehw_ is now known as ehw [10:11] any ideas why python and apt would be throwing these errors only when executing commands from within an install hook? http://paste.ubuntu.com/662523/ [11:29] <_mup_> Bug #823866 was filed: ec2metadata fails with traceback within ensemble nodes < https://launchpad.net/bugs/823866 > [12:43] <_mup_> Bug #823898 was filed: When using Ensemble, add-apt-repository no longer functions properly < https://launchpad.net/bugs/823898 > === otubo[AFK] is now known as otubo [13:04] adam_g: re bug 816169 , if you run add-apt-repository with 'LANG=C add-apt-repository ...' does that correct the issue? [13:04] <_mup_> Bug #816169: When using Ensemble, add-apt-repository no longer functions properly

< https://launchpad.net/bugs/816169 > [13:08] SpamapS: actually, launching a test that messes with the env. right now. willl let you know [13:15] cool [13:15] Daviey: I'm heading down to get some breakfast, and then maybe we can have a quick google hangout / skype chat to figure out a plan for making use of the machines. [13:16] SpamapS: same [13:22] SpamapS: sure thing === ahs3` is now known as ahs3 [13:33] SpamapS: same errors using the exact environment i'd use running it manually [13:42] adam_g: not sure I follow you exactly [13:42] adam_g: did you duplicate the hook environment locally and you got that error? [13:47] SpamapS: no, the other way around. duplicated users environment in the hook [13:49] adam_g: so may be related to the lack of controlling terminal maybe [13:55] SpamapS: strange. works fine on natty and was working on oneiric up until last week sometime [14:14] fwereade: ping [14:15] RoAkSoAx: pong [14:15] RoAkSoAx: call ran over, I'll be down shortly [14:15] fwereade: sure, you still want the branch against ensemble? [14:22] RoAkSoAx: don't worry, just in future [14:22] RoAkSoAx: it's not such a huge diff I can't handle it ;) [14:23] fwereade: heh, it is really no problem to upload it lol: lp:~andreserl/ensemble/shadow-trunk-sprint [14:23] fwereade: doesn't have any new stuff btw [14:24] RoAkSoAx: cheers [14:28] <_mup_> ensemble/states-with-principals r324 committed by kapil.thangavelu@canonical.com [14:28] <_mup_> resurrect missing security rules module [14:32] +def shutdown_ensemble(provider, kill_machines): [14:32] + return kill_machines(provider) [14:32] !!! [14:35] niemeyer: ensemble.providers.orchestra.instance is nearly as bad [14:35] * fwereade hangs head [14:36] niemeyer: that branch started off as a shutdown branch, at which point the shutdown function included the state-clearing stuff [14:36] fwereade: The whole direction of these alias/helper functions feel a bit dubious.. I'll send a review and we can talk afterwards [14:36] niemeyer: cool [14:37] Daviey: talking with RoAkSoAx .. I'm going to try and help him get on the hardware network and get one of the boxes setup w/ the latest orchestra server... which box should he use? [14:48] SpamapS: woah, wait [14:48] SpamapS: We have done that. [14:48] Or i believe we have [14:49] cmpedak is the cobbler server. [14:49] cempedak* [14:49] Daviey: cobbler or orchestra? [14:49] ;) [14:50] SpamapS: I installed orchestra-provisioning-server on Monday [14:50] adam_g re-installed it this morning, documenting the steps. [14:50] adam_g: what did you isntall [14:50] RoAkSoAx: have you made significant changes since then? [14:50] ? [14:51] I know they've been working on stuff to make it more automatic and easier to integrate w/ ensemble. [14:51] Been hard to iterate until today because our downstream bandwidth is crap [14:52] the steps i needed to take in the lab were to get dhcp and dns working properly [14:53] the only changes to orchestra/cobbler were to modify the kernel options passed via pxe [14:54] to look something like http://paste.ubuntu.com/662709/ [14:56] Ahh thats exactly what kirkland and RoAkSoAx were working on yesterday.. the DNS and DHCP [14:56] Its possible I'm just adding noise to a situation that is happening naturally. [14:57] i think the assumption is that tthese services are there and working ahead of time. they were in our case, but we had to put the orchestra server elsewhere and they weren't reconfigured correctly [14:58] adam_g: that's the simplest thing ever [14:58] that's all controlled by cobbler [14:59] adam_g: but we are currently working on getting the installation working here with the bandwidth limitations we have [14:59] ? [14:59] * hazmat yawns [14:59] adam_g: DNS/DHCP [15:06] <_mup_> ensemble/security-policy-rules-redux r332 committed by kapil.thangavelu@canonical.com [15:06] <_mup_> resolve conflict from states-with-principals [15:11] RoAkSoAx: oh, good to know [15:13] RoAkSoAx: but surely that'll require some manual steps to get systems entered into a new setup? if instead we use pre-existing servers, we should be able to just install the orchestra server and begin netbooting to automatically enroll [15:17] <_mup_> ensemble/expose-provider-ec2 r332 committed by jim.baker@canonical.com [15:17] <_mup_> Revised error messages [15:47] <_mup_> ensemble/security-policy-rules-redux r333 committed by kapil.thangavelu@canonical.com [15:47] <_mup_> update tests in aftermath of removing service state security group accessor. === jamespage1 is now known as jamespage [16:38] <_mup_> ensemble/pythonpath-fix-bug-816264 r302 committed by kapil.thangavelu@canonical.com [16:38] <_mup_> yank superflous env variable PYTHON [17:19] <_mup_> ensemble/expose-provider-ec2 r333 committed by jim.baker@canonical.com [17:19] <_mup_> Add testing around a security group is still active and cannot be deleted [17:23] <_mup_> ensemble/expose-provider-ec2 r334 committed by jim.baker@canonical.com [17:23] <_mup_> PEP8 & PyFlakes [18:04] <_mup_> ensemble/expose-provision-machines r297 committed by jim.baker@canonical.com [18:04] <_mup_> Merged trunk [18:04] <_mup_> ensemble/expose-provision-machines-reexpose r306 committed by jim.baker@canonical.com [18:04] <_mup_> Merged upstream expose-provision-machines [18:09] Our Kanban is updating correctly again [18:09] I mean, the official one, rather than the one at people.canonical.com [18:09] Please ping me if it goes crazy again [18:11] <_mup_> ensemble/pythonpath-fix-bug-816264 r303 committed by kapil.thangavelu@canonical.com [18:11] <_mup_> avoid exposing pythonpath to hooks [18:12] <_mup_> ensemble/expose-provider-ec2 r335 committed by jim.baker@canonical.com [18:12] <_mup_> Merged upstream & resolve conflicts [18:16] <_mup_> ensemble/expose-provider-ec2 r336 committed by jim.baker@canonical.com [18:16] <_mup_> PEP8/PyFlakes/leftover text conflict [18:24] robbiew: Do we have consensus around the new name for lp:principia? [18:24] robbiew: Maybe we should get together here in a moment to figure this out and move the renaming on [18:24] niemeyer: I think so...sur [18:24] sure [18:24] * robbiew digs for the email [18:33] seeing a problem on this test: ensemble.hooks.tests.test_invoker.TestCompleteInvoker.test_relation_get_format_shell_bad_vars [18:33] negronjl, kim0: note to change hadoop repo links in blog posts [18:34] m_3: will do thx for the reminder :) [18:36] negronjl: have the dummy packaging trick working... we can push to principia repos that don't already have lp src packages [18:36] m_3: that's cool. you know ... sharing is caring :) [18:37] negronjl: ha!... yeah, I'll add it to principia-tools [18:37] m_3: perfect! [18:37] m_3: thx [18:38] np [18:41] robbiew: Do you have a link for the conversation about the logo at hand? [18:41] robbiew: I'd like to reply to the request about it, but don't know what's the best thread/bug about it at that point [18:42] niemeyer: hmm...there was a bug, but I believe they already have that...one sec, let me look [18:42] robbiew: Hmm.. they've requested the logo/conversation about that, so I suspect that if they have it, they forgot abou t it [18:43] niemeyer: https://lists.ubuntu.com/archives/ensemble/2011-July/000216.html [18:43] there's the start of the thread [18:43] I'll get the bug, one sec [18:44] niemeyer: https://bugs.launchpad.net/ubuntu-branding/+bug/807100 [18:44] <_mup_> Bug #807100: Develop Ensemble logo (ensemble.ubuntu.com) < https://launchpad.net/bugs/807100 > [18:44] robbiew: Cheers! [18:44] ;) [18:46] robbiew: Sent, you're ccd [18:46] Aram: ping [18:46] cool [18:46] <_mup_> ensemble/pythonpath-fix-bug-816264 r304 committed by kapil.thangavelu@canonical.com [18:46] <_mup_> remove debug/error reproducer from example formula [19:37] there will be a momentary blip in trunk as i merge in 3 branches related to provisioning. the context is that if you're using the ec2 provider (so not testing or working on orchestra), there will not be an available impl of the security group mgmt [19:38] it should be about 15 minutes to go through the 3 merges, starting in about 5 minutes or so [19:49] <_mup_> ensemble/trunk r307 committed by jim.baker@canonical.com [19:49] <_mup_> merged expose-provision-machines [r=hazmat,niemeyer][f=809007] [19:49] <_mup_> Modifies the provisioning agent to manage the firewall on a machine. [19:49] this process has begun [19:50] never mind, 3 commits, one push [19:50] this is a happier process [19:51] <_mup_> ensemble/trunk r308 committed by jim.baker@canonical.com [19:51] <_mup_> merged expose-provision-machines-reexpose [r=hazmat,niemeyer][f=812619] [19:52] <_mup_> Modifies the provisioning agent to enable re-exposing of a service. [20:03] <_mup_> ensemble/trunk r309 committed by jim.baker@canonical.com [20:03] <_mup_> merged expose-provider-ec2 [r=niemeyer,fwereade][f=767420] [20:03] <_mup_> Modified EC2 provider to support firewall management. === otubo is now known as otubo[AFK] [20:08] firewall mgmt is now in trunk! [21:26] <_mup_> ensemble/formula-state-with-url r307 committed by kapil.thangavelu@canonical.com [21:26] <_mup_> ec2 storage has a get_url method that returns a shareable signed url for any resource. [21:44] kim0, we will need to update the include tutorial on writing a formula for drupal to take in account the new support for firewall mgmt [21:44] kim0, i'm currently going through the docs now in light of these changes having landed in trunk === otubo[AFK] is now known as otubo [21:59] <_mup_> ensemble/formula-state-with-url r308 committed by kapil.thangavelu@canonical.com [21:59] <_mup_> get url for dummy provider storage [22:01] <_mup_> ensemble/formula-state-with-url r309 committed by kapil.thangavelu@canonical.com [22:01] <_mup_> get url for orchestra file storage === otubo is now known as otubo[AFK] [22:47] <_mup_> ensemble/expose-docs r310 committed by jim.baker@canonical.com [22:47] <_mup_> Initial commit [22:55] <_mup_> Bug #824218 was filed: Update docs on EC2 firewall mgmt < https://launchpad.net/bugs/824218 > [23:02] <_mup_> Bug #824219 was filed: ensemble shutdown and terminate-machine should delete associated security groups < https://launchpad.net/bugs/824219 > [23:07] <_mup_> Bug #824222 was filed: ensemble bootstrap should be more robust < https://launchpad.net/bugs/824222 >