19:00:24 <catherineD> #startmeeting refstack 19:00:25 <openstack> Meeting started Mon Aug 31 19:00:24 2015 UTC and is due to finish in 60 minutes. The chair is catherineD. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:00:26 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:00:29 <openstack> The meeting name has been set to 'refstack' 19:00:40 <catherineD> roll call 19:00:48 <pvaneck> o/ 19:01:06 <dliu> Hello 19:01:14 <sslypushenko__> o/ 19:01:35 <catherineD> #link meeting agenda and notes, please feel free to add items https://etherpad.openstack.org/p/refstack-meeting-15-08-31 19:02:01 <catherineD> hello everyone ... 19:02:01 <Rockyg> o/ 19:02:13 <catherineD> let's start ... 19:02:17 <sslypushenko__> catherineD hi! 19:02:53 <dliu> Hello Catherine 19:02:54 <catherineD> first of all, I want to introduce dliu: ...who is from IBM China ... 19:03:00 <pvaneck> hi dliu 19:03:11 <pvaneck> welcome 19:03:14 <dliu> Hi pvaneck 19:03:27 <dliu> thanks a lot 19:03:30 <sslypushenko__> Hi, dliu! 19:03:40 <catherineD> dliu: will join us on the RefStack project ... 19:03:41 <dliu> Hi sslypushenko 19:03:41 <Rockyg> Welcome. Sorry for the early meeting... 19:03:53 <dliu> It's ok 19:04:07 <dliu> my pleasure to meet all you guys 19:04:16 <catherineD> dliu: what is your time now? 19:04:17 <dliu> Hi Rockyg 19:04:36 <dliu> 3:00 in the morning 19:04:39 <catherineD> wow 19:04:45 <dliu> :-) 19:04:57 <catherineD> dliu: sorry ... 19:05:03 <dliu> Iti 19:05:11 <sslypushenko__> really early) 19:05:14 <dliu> It's ok, don't worry 19:05:29 <catherineD> #topic No RefStack IRC next Monday (September 7) -- US holiday 19:05:47 <catherineD> sslypushenko__: and it is really late for you? 19:06:02 <sslypushenko__> not so much 19:06:13 <catherineD> #topic Relocate RefStack Project 19:06:40 <catherineD> we are now in "Big Tent" .. 19:06:53 <sslypushenko__> That is great 19:07:03 <pvaneck> :) 19:07:23 <dliu> cool 19:07:35 <catherineD> yep ... the next step is to move the repos ... I will look into creating a patch for that if needed 19:07:54 <catherineD> #topic Infra hosting 19:08:05 <Rockyg> I think Ops will handle that, but not positive 19:08:17 <catherineD> Rockyg: that would be good .... 19:08:36 <catherineD> pvaneck: any status on your patch ? 19:08:39 <Rockyg> I'll ping them on the infra channel... 19:08:43 <pvaneck> #link https://review.openstack.org/#/c/214950/ 19:08:50 <pvaneck> is the puppet-refstack patch that is needed 19:08:52 <catherineD> Rockyg: thank you .... 19:09:02 <pvaneck> it has one +2 19:09:24 <pvaneck> and is just awaiting another 19:09:28 <catherineD> pvaneck: should we get on their IRC meeting ? 19:09:45 <Rockyg> That's tomorrow at noon PDT 19:10:01 <Rockyg> But, yeah. That would be good 19:10:10 <pvaneck> i've brought it up already with jim 19:10:18 <pvaneck> should hopefully have a review soon 19:10:20 <sslypushenko__> We need to speedup hosting process 19:10:55 <Rockyg> sslypushenko__, good luck with that ;-) 19:11:03 <catherineD> Let's get on infra IRC if there is no progress ... 19:11:29 <pvaneck> still need to update the puppet-refstack refstack.conf template to account for new additions and changes 19:11:55 <catherineD> pvaneck: so we need one more patch? 19:12:06 <Rockyg> Can you have that done by noon tomorrow? I'll go see where their meeting agenda is.... 19:12:24 <pvaneck> catherined: yes, I will submit one 19:12:59 <pvaneck> yea, updating the refstack.conf should be straightforward 19:13:12 <catherineD> #action pvaneck: to submit a patch to update puppet-refstack restack.conf 19:13:46 <catherineD> Rockyg: do we need to add ourself to the infra agenda 19:14:10 <Rockyg> I'm checking now... 19:14:16 <catherineD> Rockyg: thx .. 19:14:48 <catherineD> with that can go the next topic which is kind of related to this one ... 19:15:10 <catherineD> #topic Disable anonymous data upload now that user can decide to share data anonymously 19:15:25 <catherineD> I really would like to disable that when we move to infra .... 19:16:01 <Rockyg> ++ 19:16:27 <catherineD> with anonymous data upload we can not delete data .... 19:16:32 <sslypushenko__> +1 19:16:56 <pvaneck> a key thing we need is instructions on the import key page for directions on signing keys (i.e. use refstack-client) 19:17:02 <pvaneck> i will create a bug 19:17:08 <sslypushenko__> catherine we can add admin role with right to delete public data 19:17:29 <catherineD> sslypushenko__: the problem is we do not know which data to delete ... 19:17:50 <sslypushenko__> but anyway there is not reason to keep anonymous upload 19:17:50 <catherineD> or we have no criteria on which data should be deleted .... 19:17:58 <catherineD> sslypushenko__: ++ 19:18:27 <sslypushenko__> pvaneck You are 100% right, we need updated docs 19:18:31 <catherineD> since we all agree here ... let's submit a patch to disable anonymous upload ... 19:19:01 <catherineD> #action Disable anonymous data upload to RefStack 19:19:24 <sslypushenko__> catherineD Lets update docs first 19:19:36 <catherineD> sslypushenko__: agree 19:20:52 <catherineD> #agreed Anoymous data upload should only be disabled after RefStack docs are updated with instruction for key importing 19:21:19 <catherineD> anything else? ready for the next topic? 19:21:35 <sslypushenko__> lets roll 19:21:41 <catherineD> #topic Vendor registration process 19:22:22 <catherineD> Vendor registration and comparing data are the two items left from the Vancouver f2f meeting action items ... 19:22:39 <catherineD> I have asked for a specification on the DefCore IRC meeting ... 19:23:05 <catherineD> DefCore asked that I send a email to the DefCore ML ... 19:23:12 <catherineD> I will send that email out today ... 19:23:55 <hogepodge> o/ 19:24:06 <catherineD> the specification should come from DefCore and the Foundation (hogepodge: ) 19:24:07 <sslypushenko__> catherineD I think vendor registration should be considered as out of scope for current cycle 19:24:13 <catherineD> hogepodge: just in time .... 19:24:44 <sslypushenko__> Spec for that item will be enought 19:24:52 <catherineD> hogepodge: we are discussing vendor registration process ... 19:25:46 <catherineD> sslypushenko__: yea ... at lease we should get a spec for that to plan our work ... 19:26:16 <catherineD> to prepare for that .... let's get to the next item 19:26:23 <Rockyg> If we can get the requirements/spec through the DefCore committee, that would be a great accomplishment. That is for Liberty 19:26:45 <catherineD> Rockyg: sslypushenko__: I think so .... 19:26:55 <sslypushenko__> Rockyg Fully agreed 19:26:57 <hogepodge> It's not critical short term. 19:27:10 <hogepodge> I trust that vendors will send valid test results we can link to externally 19:28:52 <catherineD> #agreed Target for vendor resgistration for the Liberty cycle is to mergte a DefCore requirement specification patch 19:29:40 <catherineD> ok next discuss the next item ... 19:29:54 <catherineD> #topic User/public key/result data relationship discussion 19:30:18 <catherineD> to prepare for vendor registration ... we need to review what we implemented today ... 19:31:00 <catherineD> today, data is associated to the key not user ... 19:31:24 <catherineD> if the key is deleted the data is zombie ... 19:31:39 <sslypushenko__> not 100% right 19:31:57 <dliu> that means we need to change the db table? 19:31:57 <catherineD> sslypushenko__: please explain ... 19:32:16 <sslypushenko__> if user upload this key back test results will be associated back 19:32:54 <catherineD> sslypushenko__: I am thinking of in the case when the user no longer works for a vendor ... 19:33:08 <catherineD> and that key belongs to the user ... 19:33:16 <sslypushenko__> хм... 19:33:20 <sslypushenko__> hmmm) 19:33:21 <catherineD> how would the vendor keep the data? 19:33:57 <sslypushenko__> It need some time to think 19:35:00 <catherineD> sslypushenko__: yea I think so .... we need to consider the relation ship among user, data and vendor .... 19:35:00 <sslypushenko__> Basically, I think that vendor will have special accoutn 19:35:26 <sslypushenko__> so vendor will act as user 19:35:50 <sslypushenko__> maybe with some additional extra features 19:36:37 <Rockyg> So, what happens if the vendor decides to change his key? The person assigned to do the uploads leaves, and the vendor for security, changes the keys..... 19:36:45 <sslypushenko__> we can add possibility to move test results from user account to vendor one 19:36:45 <catherineD> maybe there should be a master key that is owned by the vendor (master user) 19:37:40 <sslypushenko__> sure... vendor should have a specific key 19:38:20 <catherineD> sslypushenko__: Rockyg: I am thinking that each set of data will belongs to 2 keys (user key and vendor key) 19:38:47 <sslypushenko__> Sounds not bad) 19:39:09 <catherineD> when the user leaves the company, vendor will delete the user key and add outher use key if needed ... 19:39:37 <catherineD> I know we are discussiong all of this before we have a spec .. 19:39:56 <Rockyg> sounds reasonable. But perhaps we should invite the security team in OpenStack to weigh in? 19:40:09 <catherineD> but I would like is to think about these scenario and match it to the way we implemented today ... 19:40:28 <Rockyg> They might know industry standards and opensource SW for the keys/management/etc 19:40:40 <catherineD> Rockyg: absolutely ... for now I would like all of us to give some thoughts on the topic ... 19:41:00 <Rockyg> Sounds good 19:41:33 <sslypushenko__> We already have anought 19:41:40 <catherineD> #info RefStack team to review the relationship among user/data/vendore 19:41:44 <sslypushenko__> *eought thing to think 19:42:03 <catherineD> sslypushenko__: I know I know ... 19:42:17 <catherineD> that bring us to the next topic .... 19:42:24 <catherineD> #Open discussion 19:42:35 <catherineD> #topic Open discussion 19:42:55 <Rockyg> Ah. So, looks like the project teams files the patch to move the repos 19:43:20 <Rockyg> example: https://review.openstack.org/200730 19:43:23 <catherineD> Rockyg: that is super ... not thing for us to do .... thx for checking 19:44:07 <Rockyg> Once filed, Ifra adds it to its list of Project renames, which are separate from the less heavy repo moves 19:44:36 <catherineD> Rockyg: that definitely is more efficient ... 19:44:55 <catherineD> sslypushenko__: so our speaker session is not selected :-( 19:45:02 <Rockyg> once approved, it gets scheduled by Ops. The repo move is scheduled for 10/17. I don't know if they'd do projects first or at the same time. 19:45:06 <sslypushenko__> I know( 19:45:49 <sslypushenko__> Hope next time will be luckier ) 19:46:15 <sslypushenko__> But what about f2f? 19:46:18 <catherineD> sslypushenko__: I hope so ... I was hoping for us to demo all the new features 19:46:33 <Rockyg> Next time, I think the foundation will do it as one of the main sessions.... 19:46:40 <Rockyg> Or at least demo it as such. 19:46:45 <Rockyg> Maybe even this time. 19:47:19 <catherineD> Rockyg: if they do that we can recommend sslypushenko__: ? 19:47:49 <Rockyg> I would think so... as the demo runner... 19:47:58 <catherineD> anyway I was having high hope to demo to the community ... 19:48:05 <sslypushenko__> catherineD wow wow)) not so fast) 19:48:32 <sslypushenko__> I think I can't be a good demo runner) 19:48:44 <catherineD> sslypushenko__: why not? 19:49:36 <sslypushenko__> My English is really far from good enougth) 19:50:40 <catherineD> sslypushenko__: that would not be too much problem if we are demoing to a technical committee ... 19:51:02 <sslypushenko__> May be) 19:51:34 <catherineD> sslypushenko__: no pressure ... just that you have done such an amazing contribution to the project ... 19:51:50 <catherineD> back to the f2f ... 19:51:57 <sslypushenko__> But I really prefer to take part in Austin summit) 19:52:19 <sslypushenko__> If I will have a chance) 19:52:38 <catherineD> one of the major topic for the f2f is to discuss vendor registration ... 19:53:06 <catherineD> the other topic is non-tempest testing ... 19:53:44 <catherineD> for the vendor registration we just decide that the target for Liberty is a spec ... 19:53:59 <catherineD> so the only topic is non-tempest testing .... 19:54:30 <sslypushenko__> I can share some results related with this topic 19:54:32 <catherineD> Rockyg: I am thinking of inviting Randy Bias's team for t he EC2 testing ... 19:54:52 <sslypushenko__> https://review.openstack.org/#/c/214571/ 19:54:55 <Rockyg> Sounds good. Only need one or two, but that makes it easier. 19:55:39 <catherineD> sslypushenko__: that is great .... Let me check the date for a f2f with Randy's people and maybe checking with some RackSpace peole too .. 19:55:40 <sslypushenko__> It is results of work on running swift tests using tempest 19:56:06 <sslypushenko__> It is far away from ready solution 19:56:25 <catherineD> #action catherineD: to select a date for a one day f2f meeting for non-tempest testing in RefStack /// 19:56:33 <sslypushenko__> but as proof of concept it is working 19:56:48 <catherineD> sslypushenko__: that is great ... will take a look ... 19:56:59 <catherineD> anything else before we close? 19:57:22 <sslypushenko__> so we cancel f2f this week& 19:57:26 <sslypushenko__> ? 19:57:31 <Rockyg> just we need someone to write the repo move patch 19:57:49 <Rockyg> I *might* be able to do it.... 19:57:54 <catherineD> sslypushenko__: yes we decided last week that we will reschedule ... 19:58:04 <sslypushenko__> okay 19:58:21 <Rockyg> But pretty easy for pvaneck or sslypushenko__ 19:58:23 <catherineD> sslypushenko__: I hope we did not mess you up ... 19:58:33 <sslypushenko__> Me too) 19:58:37 <catherineD> on the reschedule ... 19:59:23 <catherineD> Rockyg: go for it ... 19:59:35 <catherineD> let's end the meeitng .. 19:59:53 <catherineD> #endmeeting