12:00:55 <david-lyle> #startmeeting Horizon
12:00:56 <openstack> Meeting started Wed Mar 16 12:00:55 2016 UTC and is due to finish in 60 minutes.  The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:00:57 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
12:00:59 <openstack> The meeting name has been set to 'horizon'
12:01:02 <robcresswell> o/
12:01:17 <tsufiev> hi
12:01:52 <r1chardj0n3s> o/
12:02:27 <doug-fish> \o
12:03:06 * david-lyle tries to fight through the crowd to the front
12:03:15 <mrunge> o/
12:03:29 <david-lyle> Let's get started
12:03:42 <david-lyle> RC-1 should happen this week
12:03:57 <david-lyle> that's my primary concern right now
12:04:21 <david-lyle> I've moved one bp out of RC-1 as it wasn't going to close
12:04:35 <david-lyle> The consistency group one is also at risk
12:04:53 <david-lyle> and the launch instance one needs a little more work
12:04:57 <robcresswell> I don't think volumes is going to make it; still have a concern with some parts, and there's been no other core reviews on it.
12:05:12 <robcresswell> Ports I'm just updating, that should be good to go in a couple hours
12:05:53 <david-lyle> robcresswell: ok thanks
12:06:25 <david-lyle> so I think one bp realistically is still in play
12:06:38 <david-lyle> then the remaining bugs
12:07:46 <david-lyle> there's only one or two bugs in the RC list that look worthy of slowing the RC for
12:09:18 <doug-fish> I assume you are thinking https://bugs.launchpad.net/horizon/+bug/1543891 ?
12:09:19 <openstack> Launchpad bug 1543891 in OpenStack Dashboard (Horizon) "Launch instance from Volumes Snapshot Page opens LEGACY launch instance, even if LEGACY is set to False in local_settings" [High,In progress] - Assigned to Matt Borland (palecrow)
12:09:46 <tsufiev> I would say every one of third High bugs is nice to see fixed in M
12:09:53 <tsufiev> *of three
12:10:09 <robcresswell> https://bugs.launchpad.net/horizon/+bug/1553314 This one is quite important and a small fix.
12:10:10 <openstack> Launchpad bug 1553314 in OpenStack Dashboard (Horizon) "Attempting to create a volume after deleting one fails" [High,In progress] - Assigned to Timur Sufiev (tsufiev-x)
12:10:28 <david-lyle> yes those two
12:10:59 <tsufiev> haven't looked at Angular actions after jQuery refresh recently, does it work already?
12:11:26 <tsufiev> I mean https://bugs.launchpad.net/horizon/+bug/1514627
12:11:28 <openstack> Launchpad bug 1514627 in OpenStack Dashboard (Horizon) "Angular actions not evaluated properly when dependent on row update." [Medium,In progress] - Assigned to Tyr Johanson (tyr-6)
12:11:37 <robcresswell> No, and Tyr has only put up a partial fix and no others.
12:11:38 <doug-fish> there is a patch merged I think
12:11:40 <doug-fish> oh
12:11:54 <david-lyle> what views are actually impacted?
12:12:15 <david-lyle> just images? or objects as well
12:12:24 <robcresswell> The angular actions one?
12:12:31 <david-lyle> sorry, yes
12:12:39 <tsufiev> every table with Update Metadata Actions, every table with Launch Instance NG action
12:12:44 <robcresswell> Anything that can do launch instance (volumes, images, snapshots)
12:13:01 <david-lyle> ok, that is a high then
12:13:02 <david-lyle> :(
12:13:02 <robcresswell> If oyu create one, then try and launch an instance, the button doesnt work.
12:13:15 <tsufiev> I invented an ugly ad-hoc solution for that issue: https://review.openstack.org/#/c/289849/1
12:13:25 <tsufiev> if we need some desperate measures :)
12:14:04 <robcresswell> No, that involves undoing the strict-di bp from earlier
12:14:21 <david-lyle> so what did the merged patch help, anything?
12:14:27 <tsufiev> yep, that's why I called them desperate
12:14:48 <robcresswell> Something to do with images, which isnt even enabled
12:14:59 <david-lyle> super
12:15:08 <david-lyle> so that one is not in progress
12:15:25 <david-lyle> looks like we have an opportunity
12:16:00 <david-lyle> moved to unassigned and confirmed
12:16:13 <david-lyle> please take if you have an idea
12:16:21 <tsufiev> robcresswell: but if we could find a way to inject $injector into legacy JS without undoing ng-strict-di, my solution would be less ugly
12:16:41 <robcresswell> Sure
12:16:59 <robcresswell> The thing is, there are 2 outstanding bugs against LI behaviour that I would consider blocking issues
12:17:01 <r1chardj0n3s> I'm having trouble following that, I'll have to look into it tomorrow ;-)
12:17:43 <david-lyle> so the other option is go back to legacy, which isn't ideal
12:17:46 <robcresswell> If we can't get these sorted in the next day, we'll likely have to consider reverting LI-default.
12:17:49 <robcresswell> Yes.
12:18:01 <tsufiev> oh
12:18:05 <david-lyle> let's see if we can make progress
12:18:07 <tsufiev> :(
12:18:43 <david-lyle> are there any other issues that people want to mention leading into the RC?
12:18:57 <robcresswell> I'll reviews tsufievs patch for https://bugs.launchpad.net/horizon/+bug/1553314 and get Ports updated, then see what I can do about the other bugs.
12:18:58 <openstack> Launchpad bug 1553314 in OpenStack Dashboard (Horizon) "Attempting to create a volume after deleting one fails" [High,In progress] - Assigned to Timur Sufiev (tsufiev-x)
12:19:07 <robcresswell> Oops, sorry for speaking over you there :)
12:19:22 <david-lyle> not over me, you're good
12:19:41 <david-lyle> the other thing is please continue to test
12:19:59 <david-lyle> if we find other issues we have RC-2 to correct if need be
12:20:37 <tsufiev> actually this one https://bugs.launchpad.net/horizon/+bug/1528465 could be annoying as well
12:20:40 <openstack> Launchpad bug 1528465 in OpenStack Dashboard (Horizon) "dashboard project network column display duplicate default public network randomly (with admin)" [Medium,In progress] - Assigned to Akihiro Motoki (amotoki)
12:20:43 <tsufiev> I can confirm it
12:21:18 <robcresswell> tsufiev: Patch is in the gate
12:21:22 <robcresswell> Seems its already fixed?
12:21:41 <david-lyle> yeah that one is in the gate
12:21:42 <tsufiev> robcresswell, never mind then :)
12:21:59 <tsufiev> need new status for LP: Patch is in gate
12:22:42 <david-lyle> patch is 3/4 through gate and no failing tests
12:22:56 <robcresswell> don't jinx it
12:22:56 <david-lyle> would be a helpful status in LP :)
12:23:58 <david-lyle> ok, that's the RC
12:24:08 <david-lyle> #topic summit space
12:24:50 <david-lyle> I was informed there may potentially be extra fishbowl or working session space at the summit
12:25:04 <david-lyle> do we desire any additional space?
12:25:19 <david-lyle> we were 1 fishbowl and 7 working session
12:25:24 <r1chardj0n3s> it'd be really good if we could get a cross-project yelling space to sort out xstatic ...
12:25:40 <david-lyle> that's a different pool generally
12:25:46 <r1chardj0n3s> we got *so* close last week
12:25:49 <r1chardj0n3s> ok
12:26:11 <tsufiev> is the stable/mitaka branch cut with with the first RC or with the final one?
12:26:12 <david-lyle> if we use one of our times I think we'll have lots of people with conflcts
12:26:19 <david-lyle> tsufiev: first one
12:26:29 <doug-fish> have we started collecting topics for the summit yet?
12:26:37 <david-lyle> doug-fish: not yet
12:26:51 <david-lyle> we should probably get the etherpad rolling
12:27:07 <david-lyle> I can post today
12:27:19 <doug-fish> Might be easier to sort out space needs if we knew the topics. My suspicion is that 8 is plenty, but a topic list might show otherwise
12:27:51 <tsufiev> jfyi: I won't be in Austin this summit :/
12:28:08 <doug-fish> :-(
12:28:10 <robcresswell> :(
12:28:15 <r1chardj0n3s> :( tsufiev
12:28:49 <david-lyle> ?
12:29:41 <david-lyle> well then worse performance is back on the agenda
12:29:52 <david-lyle> ;-)
12:29:59 <david-lyle> sorry to hear that tsufiev
12:30:08 <tsufiev> just more tight budget this time, hope it's the only time I was notified too late to apply for Travel Support program
12:30:36 <david-lyle> that's unfortunate
12:31:09 <david-lyle> sounds like we're ok on space
12:31:19 <david-lyle> moving on
12:31:20 <tsufiev> will use the free time to finish OSProfiler work, there are always side benefits :)
12:31:44 <david-lyle> tsufiev: a week of uninterrupted work isn't bad
12:32:27 <david-lyle> #topic PTL election
12:32:39 <tsufiev> seems that Rob has no competitors :)
12:32:46 <david-lyle> we're in the final 40  or so hours of nominations
12:32:57 <david-lyle> and yes Rob is the only brave soul to step up
12:33:15 <robcresswell> s/brave/stupid :)
12:33:21 <r1chardj0n3s> awesome
12:33:34 <tsufiev> robcresswell, aren't you afraid that you'll have to bear this for another 4 cycles ;)?
12:33:48 <david-lyle> tsufiev: don't scare him off
12:34:01 <tsufiev> haha, okay, I'm shutting up
12:34:58 <david-lyle> #topic Horizon and not-sanitized data from other OpenStack services - how should we handle it?
12:35:10 <david-lyle> this was the only item on the agenda
12:35:12 <tsufiev> ah, yes, that one is mine
12:35:36 <r1chardj0n3s> I think we should sneer a lot at the services
12:35:48 <robcresswell> r1chardj0n3s: +1
12:35:52 <tsufiev> we've got a bug recently with some Cinder data being displayed by Horizon - but it hadn't been filled in in Horizon, so it hadn't been sanitized
12:36:14 <doug-fish> link?
12:36:23 * tsufiev searching
12:36:24 <david-lyle> tsufiev: sanitized in what sense? embedded JS?
12:36:55 <david-lyle> or something less sinister
12:37:21 <tsufiev> oops, it's a private security bug
12:37:31 <doug-fish> ok understood
12:37:43 <tsufiev> let me describe it briefly
12:38:06 <tsufiev> one can put JS injection into Cinder db using CLI
12:38:15 <tsufiev> then it is displayed by Horizon
12:38:35 <david-lyle> this is not a horizon security bug?
12:38:35 <doug-fish> displayed  = executed?
12:39:01 <doug-fish> as in, it runs the javascript?
12:39:04 <tsufiev> doug-fish, yes
12:39:12 <r1chardj0n3s> ah, whoopsie
12:39:16 <doug-fish> sounds like a Horizon security bug to me
12:39:25 <david-lyle> doug-fish: I just don't see it
12:39:33 <tsufiev> david-lyle, we've been discussing (internally) who should sanitize it
12:39:47 <david-lyle> ah so not even upstream bug yet
12:39:57 <tsufiev> as of now this a our downstream repo bug, I'll file it as Horizon security
12:40:06 <tsufiev> sorry, should have done this in advance
12:40:24 <david-lyle> tsufiev: you can assign both cinder and horizon upstream and have those teams weigh in
12:40:30 <r1chardj0n3s> that poses a question: do we have cores who are allows to deal with secret security bugs?
12:40:32 <david-lyle> but horizon should sanitize it
12:40:35 <tsufiev> ok
12:40:41 <david-lyle> r1chardj0n3s: horizon-sec
12:40:46 <r1chardj0n3s> thx
12:40:55 <david-lyle> is a subset of horizon-core
12:41:30 <david-lyle> if there are cores interested please ping me or the next PTL depending on your timing of wanting in
12:41:46 <david-lyle> process is a bit different, so the group is limited
12:42:53 <tsufiev> will update bug's status in a couple of minutes, once I discuss it with our security guy
12:43:33 <david-lyle> tsufiev: make sure when you file it to mark as private security
12:43:52 <doug-fish> and don't put it out for normal review!
12:44:06 <robcresswell> If there's time, we should also discuss the new Swift UI.
12:44:13 <david-lyle> doug-fish: ++
12:44:25 <david-lyle> all reviews happen as attachments to the bug
12:44:41 <david-lyle> once there is a review in jenkins the bug is no longer private
12:44:55 <david-lyle> and will be set as such
12:45:24 <david-lyle> tsufiev: we have lots of areas where we sanitize data, so we must be missing one
12:45:50 <david-lyle> #topic swift views
12:45:53 <david-lyle> go robcresswell
12:47:05 <robcresswell> So the angular swift views have all merged
12:47:16 <r1chardj0n3s> (thanks everyone)
12:47:24 <tsufiev> :)
12:47:31 <robcresswell> Last I checked we hadnt quite decided what to do w/ regards to moving over.
12:47:48 <robcresswell> I'm in favour of holding the python content, but making the angular content default.
12:47:59 <david-lyle> that was my opinion as well
12:48:01 <r1chardj0n3s> I have a WIP with what I think is the most palatable approach https://review.openstack.org/#/c/293168/
12:48:06 <robcresswell> As it can't not be better.
12:48:22 <r1chardj0n3s> I didn't get to work on it today, due to needing to look into the ironic ui and another internal thing
12:48:24 <tsufiev> +1
12:49:17 <r1chardj0n3s> I hope to complete it tomorrow, though I wonder if that's not pushing the deadline a bit
12:50:37 <r1chardj0n3s> (noting that "tomorrow" is in 7 hours ;-)
12:51:19 <david-lyle> I may have time to fix up too
12:51:57 <r1chardj0n3s> if someone else wants to grab that patch and run with it while I'm asleep, be my guest.
12:52:48 <david-lyle> r1chardj0n3s: ok
12:52:59 <david-lyle> any other swifty toughts?
12:53:02 <david-lyle> *thoughts
12:53:19 <robcresswell> nothing from me, other than thanks for the work r1chardj0n3s
12:53:40 <r1chardj0n3s> lots more to do!
12:53:47 <robcresswell> and for persevering with my reviews :)
12:53:53 <david-lyle> yes thanks r1chardj0n3s
12:53:54 <r1chardj0n3s> :-)
12:54:17 <david-lyle> #topic hurgleburgler is a new horizon core
12:54:31 <r1chardj0n3s> \o/
12:54:45 <robcresswell> \o/
12:54:47 <doug-fish> well deserved
12:54:52 <tsufiev> more reviews \o/ :)
12:55:09 <robcresswell> *.scss -1
12:55:20 <robcresswell> I'm sure she has that scripted somewhere.
12:55:26 <david-lyle> so congrats to her and welcome
12:56:33 <david-lyle> #topic Open Discussion
12:56:45 <david-lyle> 4 minutes for your deepest thoughts
12:56:59 <r1chardj0n3s> isn't Horizon looking mighty pretty these days?
12:57:35 <tsufiev> r1chardj0n3s, on devstack it is :)
12:57:42 <tsufiev> *does
12:58:00 <tsufiev> next station is Scale Performance
12:58:38 <robcresswell> On that note, I think your nva calls patch is in the gate tsufiev
12:58:40 <robcresswell> nova*
12:59:09 <tsufiev> robcresswell, r1chardj0n3s: thank you for that, that's a beginning of a loong way :)
12:59:35 <tsufiev> will keep going there
13:00:16 <david-lyle> Thanks everyone let's wrap up the RC
13:00:21 <david-lyle> #endmeeting