19:02:12 <catherineD2> #startmeeting refstack
19:02:13 <openstack> Meeting started Mon Apr 13 19:02:12 2015 UTC and is due to finish in 60 minutes.  The chair is catherineD2. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:02:14 <sslypushenko__> o/
19:02:15 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:02:18 <openstack> The meeting name has been set to 'refstack'
19:02:38 <catherineD2> David is having an onboarding meeting for his new job at this time ... so he won't be able to attend this meeting
19:02:54 <Rockyg> Yay!  where'es he working?
19:03:16 <catherineD2> some startup forget the name ...
19:03:25 <catherineD2> Today's agenda: 1) design session planning 2) Location to host refstack server 3) plan for upgrading refstack-client to latest tempest   4) Latest Refstack UI 5) Use case for signed request 6) open discussion
19:04:08 <catherineD2> #topic design session planning
19:04:46 <catherineD2> Refstack design session #link https://etherpad.openstack.org/p/refstack-qa-vancouver-design-session
19:05:00 <catherineD2> I see no entry on the link ...
19:05:33 <sslypushenko__> No much info)
19:06:26 <catherineD2> The goal for this link is for us to log in the topic that we want for design session ...
19:06:32 <Rockyg> We at least need to put down what needs to be fixed in QA for Interop testing on outside clouds to work
19:06:45 <Rockyg> So, the SSL stuff?
19:07:30 <hogepodge> o/
19:07:50 <Rockyg> Maybe segmenting the tempest requirements.txt to be 1)default 2)discoverable 3)ask?
19:08:27 <catherineD2> Rockyg: SSL stuff is the signed request (topic 5 in today's meeting) or something else?
19:09:13 <Rockyg> Sorry, the remote sessions required in some of the tests
19:09:24 <catherineD2> ic ...
19:09:32 <Rockyg> The stuff that breaks a bunch that caused them to be flagged
19:10:38 <Rockyg> Another thing to discuss with Qa is some standards on API tests so that once a bunch of tests move to projects, we don't have a zillion different ways we have to setup to run them
19:10:59 <catherineD2> ic ssh ... that have several patches  going on now ...I think that should be from DefCore point of view ?
19:11:00 <Rockyg> Also functional tests, I would think
19:11:29 <Rockyg> hogepodge knows most about them.
19:11:50 <Rockyg> He can give us a status on patches and whether we need to talk more in Vancouver
19:13:15 <catherineD2> hogepodge: do we need to have a design session for the test flagged for ssh reasion?  More important, what are the topics we want to bring up ?
19:13:55 <hogepodge> catherineD2: Right now the qa team is soliciting design session topics
19:14:11 <hogepodge> We have unti lthe 30th to request items we want to discuss.
19:14:21 <hogepodge> So ssh tests would be appropriate.
19:14:25 <catherineD2> ic so there is still time
19:14:57 <hogepodge> I need to dig up the link. It's probably in the qa meeting from last week. The know it's a problem, and want to refactor in the next cycle.
19:15:06 <hogepodge> (as with all things network)
19:15:39 <catherineD2> do we need to discuss about the topic today or we again let everyone have a chance to enter their topic in the link https://etherpad.openstack.org/p/refstack-qa-vancouver-design-session
19:15:41 <hogepodge> I would strongly suggest that the refstack team start directly speaking to the openstack-qa team, though. They really like to have the 1:1 interaction.
19:15:44 <catherineD2> and revisit next week?
19:16:28 <Rockyg> hogepodge:  +1
19:16:28 <sslypushenko__> catherineD2 +!
19:17:35 <catherineD2> OK  everyone please enter the topics that you think we should bringup for design session in the link ... thx
19:18:16 <catherineD2> #agreed revisit topics for design session next week
19:19:03 <catherineD2> #action Refstack team member to enter suggested topics for design session in https://etherpad.openstack.org/p/refstack-qa-vancouver-design-session
19:19:26 <catherineD2> alright moving onto the next topic
19:19:40 <catherineD2> #topic Location to host refstack server
19:20:22 <sslypushenko__> Any ideas?
19:20:29 <catherineD2> First question, do we want to host a public Refstack server  ..
19:20:49 <hogepodge> o/
19:20:53 <hogepodge> Good news.
19:21:45 <Rockyg> waiting to hear!
19:21:47 <hogepodge> This morning I got an e-mail. Turns out that Piston transferred refstack.org to the Foundation this morning.
19:21:57 <Rockyg> Yay!
19:22:00 <catherineD2> hogepodge: wow great !!!
19:22:21 <catherineD2> so we have our domain back to us ...
19:22:41 <hogepodge> I don't know what that means for me (I don't have access to manage items in our domain accounts), but I can work directly with foundation staff to get the domain to point to an address
19:22:46 <sslypushenko__> Wow!
19:23:28 <catherineD2> I  take that we would like to host a public Refstack server then .. so we need to decide on where to host it ...
19:24:00 <catherineD2> please take a look at  http://pvaneck.github.io/#/   that is how Refstack website look like for now ...
19:25:39 <catherineD2> The "Community Results" tab will host data ... it can be sample data at summit time ..
19:25:49 <sslypushenko__> I tried to talk with my manager about idea to host public Refstack server on Mirantis Openstack Express by still haven't  got any answer
19:26:05 <catherineD2> sslypushenko__: thanks!
19:26:43 <catherineD2> also I think clee also mention to host at DreamHost ...
19:27:27 <catherineD2> hogepodge: will the foundation interest in hosting it ?
19:27:46 <hogepodge> catherineD2: I have no idea. It's outside of my scope of expertise.
19:27:56 <hogepodge> I think the best place to ask is in infra
19:28:19 <catherineD2> Rockyg: Would you be able to contact infra for us?
19:28:29 <mtreinish> why not just push patches to have it run in infra?
19:28:56 <Rockyg> I can get onto infra after the meeting and talk to them
19:29:04 <Rockyg> A VM is the way to go.
19:29:27 <catherineD2> mtreinish: thx for the suggestion .. We will look into that
19:29:36 <catherineD2> Rockyg: thx ...
19:30:11 <Rockyg> We'd need to essentially generate a contract doc that is everything we would provide and from where and what what they would provide.  Essentially, the BOM
19:30:48 <catherineD2> #action Rockyg: will contact infra team... Refstack team investigating the possibility to push a patch to host Refstack server at infra
19:31:21 <catherineD2> moving to the next topic ..
19:31:43 <catherineD2> #topic  plan for upgrading refstack-client to latest tempest
19:32:06 <catherineD2> currently refstack-client is still using tag-3
19:33:07 <catherineD2> Reviewing last week's IRC log, I am still not clear on what we have decided.  I believe we decide on not waiting for tag-4.   We know that refstack should be in sync with DefCore on the tempest version that they used to create the must pass list.
19:34:47 <Rockyg> so, I think most, if not all the projects have cut an RC-1, which means branching has occured on those projects.  I would suggest making sure all the defcore projects have achieved RC-1 then grabbing the SHA from tempest after the last one branched
19:34:48 <catherineD2> hogepodge: Rockyg: were in a DefCore capability meeting last week ... I think DefCore decide to ping on a certain tempest ... but I do not know which one
19:34:54 <sslypushenko__> We can use commit closest to date of approvement for new capabiliteis
19:35:28 <hogepodge> catherineD2: just something recent, as of around now
19:35:34 <hogepodge> later than the network fixes
19:35:39 <catherineD2> I think DefCore should decide on a SHA and then Refstack will follow ..
19:35:43 <hogepodge> and the idempotent_id tag
19:36:13 <Rockyg> Defcore meets again on Wednesday, but maybe we should just as QA to give us one they know is good?
19:36:56 <catherineD2> mtreinish: your thoughts?
19:36:58 <Rockyg> Then, if QA says it's good, and we run some tests, we can tell defcore and they can just ok it.
19:37:18 <catherineD2> Rockyg: +1
19:38:13 <catherineD2> Rockyg: The important thing is DefCore and Refstrack should be in sync ..
19:38:35 <Rockyg> catherine:D2: ++
19:38:36 <mtreinish> catherineD2: we continuously deploy tempest so anything should be fine to use
19:39:29 <Rockyg> mtreinish: any important fixes coming up, or is it pretty stable right now?
19:39:58 <mtreinish> Rockyg: we try to fast track important fixes as they come up
19:40:53 <mtreinish> Rockyg: I'm not sure what you're looking for on a specific sha1 though?
19:41:17 <Rockyg> mtreinish, hogpodge: so I'm thinking there is about to be a bunch of libraries bounded in requirements.txt.  Maybe we should wait for a good run or two after that?
19:41:49 <hogepodge> Rockyg: we're shooting for 2015.05 anyway, so it's best just to set expectations for what's there right now.
19:42:02 <Rockyg> mtreinish: lloking for a SHA1 that is known to be successful until we get to tag-4 for use.
19:42:03 <mtreinish> Rockyg: well we're in reqs freeze right now so the tempest requirements file won't change until that ends
19:42:17 <Rockyg> mtreinish: kewl
19:42:39 <mtreinish> Rockyg: well on that criteria any of them will work, because we cd the tempest repo
19:42:46 <catherineD2> #agreed Refstack will pick a SHA , running tests to verify and then inform DefCore on the SHA that they should use
19:42:52 <Rockyg> So, hogepodge, propose what the head  SHA1 is, test and and we're there.
19:43:24 <Rockyg> sound good, mtreinish?
19:43:37 <hogepodge> 7c8fcc67
19:43:43 <mtreinish> sure
19:44:10 <catherineD2> #agreed Refstack will use 7c8fcc67
19:44:29 <sslypushenko__> I can update ./setup_env script
19:44:40 <catherineD2> thx sslypushenko__:
19:44:42 <Rockyg> catherineD2, you can use that SHA1 for your tests, too.  Report that it works (or doesn't) to Defcore on Wednesday and get their blessing
19:45:30 <catherineD2> #action catherineD2: will tests 7c8fcc67 and report test result on Wednesday DefCore meeting
19:46:02 <catherineD2> sslypushenko__: please push a patch ...
19:46:15 <catherineD2> ready for the next topic ..
19:46:20 <catherineD2> ?
19:46:35 <sslypushenko__> catherineD2 ok
19:46:41 <catherineD2> #topic Latest Refstack UI
19:47:10 <sslypushenko__> Greate thx to Paul!
19:47:11 <catherineD2> Paul has worked on the UI for Refstack using AngularUI/AngularJS.  Demo page at http://pvaneck.github.io/#/   Right the content in the "Home" and "About" tabs are from the Refstack readme file.  Please review and update   if needed.
19:47:41 <Rockyg> LGTM
19:47:50 <catherineD2> hogepodge: Rockyg: please review the content on the "Home" and "About" tab
19:48:01 <vladiskuz|2> I saw Paul's patch. Looks good, but what about unit test?
19:48:06 <pvaneck> I should have some tests for it in a little bit
19:48:07 <catherineD2> Paul is working on the last tab to show data ...
19:48:13 <Rockyg> And, yeah, content needs review, plus I think DefCore should review content.
19:48:31 <catherineD2> vladiskuz|2: thx for review and suggesting for tests ...
19:48:39 <sslypushenko__> pvaneck Lets skip tests for first cycle
19:49:02 <catherineD2> Rockyg: hogepodge: could you ask DefCore to review
19:49:20 <Rockyg> catherineD2: yup
19:49:25 <hogepodge> sure
19:49:55 <catherineD2> using Paul's demo link for now ... since we do not have a public server yet ..
19:50:29 <catherineD2> #link http://pvaneck.github.io/#/
19:50:45 <catherineD2> #topic Use case for signed request
19:51:22 <catherineD2> so sslypushenko__: vladiskuz|2: has been working on the signed request ....vladiskuz|2:  will work on fetching result with key next ?
19:51:51 <sslypushenko__> I think I will do this feature
19:52:01 <catherineD2> ic thx sslypushenko__:
19:52:05 <sslypushenko__> Vlad is working on UI login now
19:52:21 <catherineD2> right now we can upload data with and withou key ...
19:52:24 <vladiskuz|2> catherineD2: I still working on auth with openstackid. And Paul's patch more important for me now
19:52:47 <catherineD2> vladiskuz|2: ic ...
19:53:06 <sslypushenko__> I think that for now signed request will help users to revisit own test results...
19:53:08 <catherineD2> so we should try to merge Paul's as soon as possible ...
19:53:47 <vladiskuz|2> But one important stuff is SSL, because openstackid works only with https
19:53:52 <catherineD2> sslypushenko__: would it be a better option if user get own results with cpid?
19:54:53 <sslypushenko__> In future, I guess that upload signed results will be a single way to upload results
19:55:19 <catherineD2> sslypushenko__: that would make sense ..
19:55:58 <catherineD2> vladiskuz|2: with openstackid ... where are the user credential saved at ?
19:56:15 <sslypushenko__> Using a digital signature can be considered as a type on auth
19:56:24 <catherineD2> we only have 5 mins ... Let's discuss this in #refstack after this meeting ...
19:56:44 <catherineD2> #topic open discussion ...
19:57:21 <catherineD2> anything any one want to bring up?
19:57:22 <vladiskuz|2> catherineD2: credential saved on openstackid server, but we can request some information from it. For example email
19:57:27 <sslypushenko__> catherineD2 In refstack client we can't use openstackid for auth
19:58:49 <catherineD2> Let me end the meeting here ... we will discuss the signed request topic in #refstack
19:59:09 <Rockyg> Thanks!
19:59:15 <catherineD2> Thank you all!  It has been very productive meetings lately!!!
19:59:27 <catherineD2> #endmeeting