16:01:00 <TheJulia> #startmeeting ironic_bfv
16:01:01 <openstack> Meeting started Thu Apr 27 16:01:00 2017 UTC and is due to finish in 60 minutes.  The chair is TheJulia. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:03 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:04 <mjturek> o/
16:01:05 <openstack> The meeting name has been set to 'ironic_bfv'
16:01:06 <TheJulia> o/
16:01:23 <derekh> o/
16:01:27 <mjturek> welcome back TheJulia
16:01:29 <hshiina> o/
16:01:48 <TheJulia> Thanks!
16:02:03 <TheJulia> #link Our Agenda, as always https://wiki.openstack.org/wiki/Meetings/Ironic-BFV
16:02:23 <TheJulia> #topic Announcements/Reminders
16:02:51 <TheJulia> I'm back from vacation and two business trips \o/
16:02:56 <TheJulia> And somehow I'm still alive
16:03:11 <TheJulia> Does anyone have anything to announce this week?
16:04:27 <mjturek> seems like a no :)
16:04:33 <TheJulia> Yup
16:04:33 <derekh> not sure if its an announce, but I'm hoping to help out here if I can
16:04:40 <TheJulia> derekh: Welcome!
16:04:46 <mjturek> welcome!
16:05:00 <TheJulia> Moving on!
16:05:02 <derekh> thanks
16:05:05 <TheJulia> #topic Current Status
16:05:22 <TheJulia> I'm a little worried because I've not had time to digest the ether pad or check the review sets this morning.
16:05:45 <TheJulia> #link https://etherpad.openstack.org/p/Ironic-BFV
16:06:15 <mjturek> Cinder driver - didn't get a ton done but I fixed a problem with failing docs tests this morning. I also introduced a patch in hopes of handling one of the TODOs there. I have it in discussion for later on
16:06:31 <TheJulia> Awesome
16:06:39 <TheJulia> I guess your the light gold color in the ether pad?
16:06:52 <mjturek> and orange
16:07:00 <mjturek> it seems to fade to gold after about a week :)
16:07:38 <TheJulia> :)
16:09:18 <TheJulia> Okay, looks like the ether pad is in decent shape, I just updated the notes on two of the revisions to be more up to date.  Looks like https://review.openstack.org/#/c/406290 needs a rebase too
16:10:00 <TheJulia> I guess that is it for status, there really is nothing else to cover it seems.
16:10:11 <TheJulia> So if there are no objections, we can move on to discuss priorities
16:10:46 * TheJulia takes the silence as agreement to proceed
16:10:52 <mjturek> +1
16:10:57 <TheJulia> #topic Planning/Priorities
16:13:09 <TheJulia> Looks like hshiina found that there was some extra validation logic inside cinder that I didn't find when I dug through the code originally.  Which seems kind of silly since it is asserting things that some drivers don't actually even use :(
16:13:18 <TheJulia> But it is enforcing the contract as it were
16:13:35 <mjturek> good finds hshiina
16:14:14 <TheJulia> So that seems like it is going to take some refactoring on our part, unless we just want to catch the error somehow and make it visible.  I think he pointed this out in the validation checks on our side, so it makes sense to store the data and just double check on our end
16:15:24 <TheJulia> The task at hand seems fairly clear actually, Anyone up for volunteering to tackle it?
16:15:53 <mjturek> TheJulia: is it just cinder driver changes?
16:16:24 <mjturek> to fix it I mean
16:16:37 <TheJulia> Yes
16:17:03 <mjturek> I'll volunteer then, but I'm planning on just catching the errors (unless someone has a better option)
16:17:28 <TheJulia> My brain needs to digest it a little more, to be honest
16:17:42 <mjturek> fair enough, we can discuss off meeting
16:17:43 <TheJulia> I think catching the errors and returning them to the "operator" makes more sense
16:17:49 <TheJulia> keeps ironic lighter weight
16:18:01 <mjturek> gotcha
16:18:19 <TheJulia> That would be a topic my brain goes back and forth on though :)
16:18:55 <TheJulia> #info mjturek to look at addressing hshiina's finds on the cinder driver with regards to required parameters for connections.
16:19:30 <TheJulia> I'll try to rebase the storage interface calls wire-in before the next meeting.  With the summit coming up, I have a few things on my plate so I'm not sure I'll get to it.
16:20:01 <TheJulia> #info TheJulia to rebase the storage_interface wire-in change.
16:20:29 <hshiina> mjturek, TheJulia thanks.
16:20:53 <TheJulia> I think that is good unless someone has plans or priorities to convey?
16:21:42 <mjturek> +1 for moving along
16:21:50 <hshiina> i agree that cinder driver is the priority
16:22:24 <TheJulia> Moving along then!
16:22:46 <TheJulia> #topic Discussion
16:23:27 <mjturek> So my first item was the TODO on line 271 here https://review.openstack.org/#/c/366197/39/ironic/drivers/modules/storage/cinder.py
16:24:26 <mjturek> TheJulia mentions that retrieving the target from volume_id is preferred
16:24:30 <mjturek> so I proposed this https://review.openstack.org/#/c/460250/
16:24:38 <TheJulia> #link https://review.openstack.org/#/c/366197/39/ironic/drivers/modules/storage/cinder.py#271
16:24:44 <mjturek> #link https://review.openstack.org/#/c/460250/
16:25:15 <TheJulia> #undo
16:25:15 <openstack> Removing item from minutes: #link https://review.openstack.org/#/c/460250/
16:25:19 <TheJulia> #undo
16:25:20 <openstack> Removing item from minutes: #link https://review.openstack.org/#/c/366197/39/ironic/drivers/modules/storage/cinder.py#271
16:26:08 <TheJulia> Well, when did that stop working :(
16:26:10 <TheJulia> #link https://review.openstack.org/#/c/366197/40/ironic/drivers/modules/storage/cinder.py
16:26:19 <TheJulia> #link https://review.openstack.org/#/c/460250/
16:26:30 <mjturek> not sure :(
16:27:24 <TheJulia> Anyway, looking at it
16:28:35 <TheJulia> Being able to query directly helps get rid of the loop on 270, which makes sense
16:29:29 <mjturek> cool, what about the fact that it returns a list of targets? The volume_id doesn't have a unique constraint so I figured it needs to be that way
16:29:58 <TheJulia> I have no objection to adding the db api call and updating that code accordingly, or it gets rid of the loop on 269.  Regardless, it becomes simpler and less looping
16:30:15 <mjturek> \o/ alright I'll rebase the driver on top of it and use it
16:30:23 <TheJulia> Yeah, that I need to think about though
16:31:23 <TheJulia> So as long as the volumes are limited to the node the task is executing on, I think it is okay
16:31:35 <TheJulia> I mean targets, not volumes
16:31:48 <TheJulia> if the targets returned are on different nodes, then that is problematic
16:31:48 <mjturek> if there's one where the node_uuid doesn't match, should it throw an exception?
16:32:14 <mjturek> okay I think that answers my question lol
16:32:18 <TheJulia> heh
16:32:28 <TheJulia> You could have multiple nodes connected to the same target though
16:32:42 <mjturek> hm
16:32:43 <TheJulia> but they should have distinct records for each node in the targets
16:35:07 <mjturek> hmm, alright I think I'm confused then. I'll need to think that over
16:35:43 <mjturek> oh I think I get it
16:35:47 <TheJulia> :)
16:36:11 <mjturek> anyway, probably a topic for the code review :)
16:36:32 <TheJulia> Yeah, Anyway, I guess we're good for set discussion topics.  Everyone ready to move along?
16:36:42 <mjturek> sure
16:37:21 <hshiina> yes
16:37:57 <TheJulia> Excellent!
16:38:02 <TheJulia> #topic Open Discussion
16:38:16 <TheJulia> I have nothing else, other than I'm glad to be home :)
16:38:23 <derekh> before I can help out on things I need to get myself familiar with whats going on,
16:38:23 <derekh> if I get all the correct patches from the etherpad should I be able to boot a instance from a volume or is there more work yet to happen?
16:38:36 <derekh> its always nice to be home ;-)
16:38:44 <derekh> Once I am familiar it, I'm happy to take suggestions on how to help out
16:42:12 <hshiina> derekh, i guess we need more work. there is a wip patch in ehterpad
16:42:40 <TheJulia> derekh: A little more work, mjturek was looking at getting things up and running in devstack. There is the tip patch that I posted to get feedback that needs to be finished
16:42:48 <mjturek> derekh: hshiina also detailed how to try out the patches with devstack. Take a look in the devstack section\
16:42:55 <TheJulia> Also, the nova changes, but hshiina and mjturek have been looking at that
16:44:37 <TheJulia> I feel like someone else was seeking background/context for BFV earlier in the week that could also be helpful for you derekh, but I'm completely drawing a blank on who pinged me.
16:45:04 <derekh> ok, I will take a look through the devstack section, to see how far I get, I'll ping should I hit any problems/questions, I only started setting up a env to try it out today so still feeling my way around
16:45:13 * TheJulia gets out ye olde jet lag sign and just holds it above her head
16:45:30 <derekh> TheJulia: it was me, I mentioned in the virt sprint that I was hoping to help out
16:45:32 <TheJulia> derekh: Awesome!
16:45:48 <TheJulia> Oh, right!
16:46:02 <TheJulia> I think I was walking into a datacenter during that conversation too :)
16:46:38 <TheJulia> derekh: well, as with anything, we are just a ping away in #openstack-ironic :)
16:46:39 <derekh> lol, anyways I can take a look at things and try them out, then I'll hopefully have something sensible to say next week
16:46:49 <TheJulia> Excellent!
16:47:33 <TheJulia> Out of curiosity, is anyone going to be in boston for the Summit?
16:47:45 <derekh> not me
16:47:53 <hshiina> i'm not going
16:48:09 <mjturek> TheJulia: I'll be there :)
16:50:11 <TheJulia> Awesome, I was just kind of curious since I'll be there as well.
16:50:47 <TheJulia> Anyway, if we have nothing else to discuss, I say we call the meeting done, and get back to our various tasks for the day.
16:51:03 <mjturek> ttyl all
16:51:15 <derekh> ttyl, thanks
16:51:28 <hshiina> thanks
16:52:25 <TheJulia> Thank you everyone!
16:52:28 <TheJulia> #endmeeting