16:01:50 #startmeeting ironic_bfv 16:01:51 Meeting started Thu Jun 8 16:01:50 2017 UTC and is due to finish in 60 minutes. The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:52 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:54 The meeting name has been set to 'ironic_bfv' 16:02:01 Good morning everyone! 16:02:15 o/ 16:02:22 Our agenda for today 16:02:25 #link https://wiki.openstack.org/wiki/Meetings/Ironic-BFV 16:02:43 #topic Announcements/Reminders 16:03:19 I believe this is the week of Pike-2, so we definitely need to start getting some reviews and some of the code landed soon. 16:03:48 that's right 16:04:17 o/ 16:04:25 hey everyone :) 16:04:26 I and mjturek rebased/updated some of the patches to try and get that cleaned up. Hopefully later today well have patches with clean ci runs ready for review 16:04:40 * TheJulia should have said that backwards, but *shrug* 16:05:02 That being said, that is also current status :) 16:05:09 yeaaah, screwed up py27/35 and pep8 in a couple places. Will have fixed shortly 16:05:40 #info Reviewers, please consult the list of revisions at https://etherpad.openstack.org/p/Ironic-BFV 16:06:06 I guess we should move on to current status unless anyone has anything to announce? 16:06:56 * TheJulia thinks she has crickets, or everyone is watching US congressional testimony 16:07:11 oh - I probably can't make next week's meeting, but I will try! 16:07:25 mjturek: Okay, good to know, thanks! 16:07:31 Moving on! 16:07:36 #topic Current Status 16:07:50 Our etherpad is fairly up to date, https://etherpad.openstack.org/p/Ironic-BFV 16:08:39 mjturek is going to fix a couple minor breaks above, and we'll get everything rebased so we have verified+1 for CI 16:08:53 I also saw the notes from last week \o/ 16:09:01 but we can discuss that more in discussion. :) 16:09:09 Anyone have any status updates to report? 16:09:12 We were talking about tempest and a new CI job last week, I've proposed a new CI job here https://review.openstack.org/#/c/470307/ 16:09:33 yep it looks good derekh! 16:09:37 derekh: thanks 16:09:47 but its a WIP until the tempest test are ready any we know what ironic-tempest-regex to use 16:10:00 oohhhh 16:10:08 Is that on the ether pad? 16:10:22 yup, line 38 16:10:40 awesome! 16:10:47 Wow, things are coming together kind of nicely 16:12:04 Anything else to bring up? 16:13:01 Moving on if there is nothing else 16:13:23 * TheJulia waits a little bit 16:13:57 nothing from me 16:13:58 I'm good - I'm pretty sure I've handled the comments in the current two patches. If I've missed anything, please ping me 16:14:07 Awesome 16:14:08 #topic Planning/Priorities 16:14:48 I think the priorities at the moment are getting the patchset in order with CI feedback and reviews. 16:15:57 The skip deployment patch may need a little more work, hshiina I left a reply in review after checking the revisions that mjturek updated. 16:16:23 I'd like to discuss that, if you have time hshiina 16:16:41 * TheJulia feels like she is just kind of stating the obvious today :) 16:17:43 yeaaah need to dig into the skip patch a bit today, will review 16:17:53 TheJulia: sure. 16:18:31 I think we're good to move on to Open Discussion at this point, unless there is anything we should incorporate or that needs to be raised? 16:19:31 I'm good :) (wow we're blazing through today) 16:19:45 *shrugs* 16:19:56 #topic Open Discussion 16:21:48 bare with me a moment, I'm looking for a specific revision to pose a question/thought 16:21:58 np 16:23:56 So, going back to https://review.openstack.org/#/c/454243/ which is the skip patch, and then glancing at https://review.openstack.org/#/c/441944/9/ironic/drivers/modules/agent.py which is the patch hshiina proposed. Which adds task.driver.boot.prepare_instance(task). I feel like that is already called by the Agent deploy driver's prepare method. 16:24:36 I think that would cover it, with the wire-in being in the conductor before deploy.prepare, but I could very well be wrong or recalling something incorrectly 16:26:22 so, looking for the link, it looks like I'm wrong 16:26:40 we would need to re-call it it seems, since we are truly only performing attachment on power ops in this case 16:27:06 err, that is only when active 16:27:43 so basically prepare_instance wasn't getting called unless the node was already ACTIVE? 16:27:51 https://review.openstack.org/#/c/406290/34/ironic/drivers/modules/agent.py 16:28:06 so yeah, we do need to still call prepare_instance because we are in DEPLOYING 16:28:39 prepare_instance only gets called because of the conductor takeover code 16:29:29 TheJulia: yes, prepare_instance is not called in bfv case. 16:29:45 I'll update the revision this afternoon to reflect that 16:30:04 TheJulia: thanks 16:30:14 cool 16:31:37 hshiina: I think i still need to digest your changes in ironic/drivers/modules/pxe.py. Do you see them as necessary with the skip patchset? 16:32:52 i think it is necessary 16:33:28 some configuration for pxe is done in prepare_ramdisk 16:34:03 if prepare_ramdisk is not called, some work seems necessary in prepare_instnce 16:35:06 Okay, that seems very likely given how complex both of those methods are 16:36:56 I'll update the skip revision later today 16:37:21 Anything else to discuss today? 16:39:06 i have nothing 16:39:58 Excellent, Thank you everyone! 16:40:05 thanks TheJulia! 16:40:07 ttyl 16:40:07 ty 16:40:12 thank you! 16:41:21 #endmeeting