| *** jmckind has joined #openstack-defcore | 00:23 | |
| *** jlk has quit IRC | 00:46 | |
| *** jlk has joined #openstack-defcore | 00:46 | |
| *** jlk has quit IRC | 00:46 | |
| *** jlk has joined #openstack-defcore | 00:46 | |
| *** cjvolzka has quit IRC | 01:23 | |
| *** cjvolzka has joined #openstack-defcore | 01:23 | |
| *** cjvolzka has quit IRC | 01:23 | |
| *** jmckind is now known as jmckind_ | 02:40 | |
| *** jmckind_ has quit IRC | 02:46 | |
| *** jmckind has joined #openstack-defcore | 03:22 | |
| *** jmckind is now known as jmckind_ | 03:29 | |
| *** jmckind_ has quit IRC | 03:35 | |
| *** jmckind has joined #openstack-defcore | 03:39 | |
| *** jmckind has quit IRC | 04:09 | |
| *** jmckind has joined #openstack-defcore | 04:39 | |
| *** pcaruana has joined #openstack-defcore | 05:36 | |
| *** pcaruana has quit IRC | 05:49 | |
| *** jmckind has quit IRC | 05:59 | |
| *** kbaikov has quit IRC | 06:29 | |
| *** kbaikov has joined #openstack-defcore | 06:42 | |
| *** pilgrimstack1 has joined #openstack-defcore | 07:09 | |
| *** pilgrimstack has quit IRC | 07:10 | |
| *** markvoelker has quit IRC | 07:58 | |
| *** pcaruana has joined #openstack-defcore | 08:02 | |
| *** markvoelker has joined #openstack-defcore | 08:58 | |
| *** edmondsw has joined #openstack-defcore | 12:46 | |
| *** markvoelker has quit IRC | 12:48 | |
| *** markvoelker has joined #openstack-defcore | 12:49 | |
| *** cjvolzka has joined #openstack-defcore | 13:57 | |
| *** pcaruana has quit IRC | 16:05 | |
| *** johnthetubaguy has quit IRC | 16:15 | |
| *** johnthetubaguy has joined #openstack-defcore | 16:15 | |
| *** catherineD|2 has joined #openstack-defcore | 16:58 | |
| catherineD|2 | o/ | 16:59 |
|---|---|---|
| markvoelker | o/ | 16:59 |
| eglute | catherineD|2 can you please repeat your last question? | 16:59 |
| *** rockyg has joined #openstack-defcore | 16:59 | |
| catherineD|2 | sure eglute | 16:59 |
| catherineD|2 | RefStack vendor registration questions. ( http://lists.openstack.org/pipermail/defcore-committee/2016-February/001035.html ) | 16:59 |
| catherineD|2 | do we allow "Vendors registering with RefStack not being in OpenStack MarketPlace" | 16:59 |
| purp | o/, but only partly due to a 9a call. | 17:00 |
| eglute | i think i am ok with vendors not being in marketplace | 17:00 |
| eglute | since it might take time for them to get into marketplace as well | 17:01 |
| markvoelker | Curious: what's the use case for a vendor going through the process to get a logo agreement but not being in the marketplace? | 17:01 |
| catherineD|2 | So at RefStack anyone can register a "private" vendor .... will need foundation approval before it becomes a official vendor | 17:01 |
| markvoelker | eglute: ah, you're thinking it's a timing thing? Gotcha... | 17:01 |
| eglute | hogepodge what do you think | 17:01 |
| eglute | markvoelker right | 17:01 |
| markvoelker | No idea how long it takes to get a Marketplace entry...hogepodge might | 17:01 |
| eglute | since right now it is a manual process | 17:01 |
| *** gema has quit IRC | 17:01 | |
| catherineD|2 | markvoelker: The use case is for general interops test like ec2 compatibility tests | 17:01 |
| *** gema has joined #openstack-defcore | 17:02 | |
| markvoelker | catherineD|2: Got it. Although today that's not something we actually support, so perhaps a bit lower priority? | 17:02 |
| eglute | right, and also for companies that want to upload their private cloud info, but it would never be listed in marketplace, since they would not be interested in selling it | 17:02 |
| catherineD|2 | I think RefStack code should facilitate interops tests but the RefStack Site should be for OpenStack trade mark | 17:03 |
| markvoelker | eglute: Do we really have people uploading data on their private clouds? What's the motivation for doing so? | 17:03 |
| eglute | catherineD|2 i agree with that | 17:03 |
| catherineD|2 | markvoelker: I think so in term of priority ... | 17:03 |
| eglute | markvoelker right now i am guessing! hogepodge, what have you seen? | 17:03 |
| markvoelker | eglute: =) I could see us potentially trying to motivate people to do so in the future to give us a better sense of what's actually supported/used "in the real world", just wondering if anyone's actually doing it today. | 17:04 |
| eglute | right... we want them to upload :) | 17:05 |
| rockyg | privat cloud info could become a thing for private clouds allowing hybrid/bursting | 17:05 |
| catherineD|2 | markvoelker: eglute: I am OK if the guidance from DefCore/Foundation is to concentrate on OpenStack official vendors (that is vendors exist in Marketplace) for now | 17:06 |
| rockyg | Also, some sites might have both private and public and want to assure users that moving between the two will work. | 17:06 |
| hogepodge | eglute: I only see results that are submitted to me | 17:07 |
| rockyg | Like governments, schools, etc | 17:07 |
| hogepodge | markvoelker: I have a list of individuals who are interested in running refstack and submitting data. We're parsing the latest survey results to finalize the list. | 17:07 |
| hogepodge | We're going to be asking them to follow up on that and submit results before Austin | 17:08 |
| rockyg | hogepodge, cool! | 17:08 |
| markvoelker | hogepodge: Cool. Non-vendor individuals I take it? So you're saying we definitely don't want to use the MarketPlace as an identity source? Look for a looser coupling between the two? | 17:09 |
| catherineD|2 | hogepodge: please have them upload data with signature ... Once we have vendor/product registration in place ... they can associate their data to it | 17:09 |
| eglute | hogepodge whats your preference regarding catherineD|2 original question? | 17:09 |
| hogepodge | markvoelker: the person who filled out the survey, be it for themselves or on behalf of their employer | 17:09 |
| hogepodge | vendors shouldn't have to be in marketplace to be listed in refstack | 17:10 |
| rockyg | hogepodge, ++ | 17:10 |
| hogepodge | catherineD|2: ^^ | 17:11 |
| eglute | catherineD|2 does that work for you? "vendors shouldn't have to be in marketplace to be listed in refstack"? | 17:11 |
| catherineD|2 | hogepodge: the vendors that are in RefStack and not marletplace are they "official" vendors? | 17:11 |
| hogepodge | catherineD|2: we should work to make sure vendors are accurately represented in refstack | 17:11 |
| rockyg | catherineD|2, so not necessarily "vendors" but certified operators? | 17:12 |
| rockyg | certified via passinf refsack. | 17:12 |
| markvoelker | So pointing back to http://lists.openstack.org/pipermail/defcore-committee/2016-February/001035.html questions 1A and 1B... | 17:12 |
| catherineD|2 | hogepodge: certify a vendor is a big job for RefStack and we are not ready for that for the Mitaka release | 17:12 |
| hogepodge | I don't have a sense of official. It can have so many definitions. To be in the marketplace you must be a sponsor. | 17:13 |
| markvoelker | I think what catherineD|2 is getting at is: if we aren't going to use the MarketPlace as an identity source, what information does RefStack need to collect/store/etc to create a "vendor entity"? | 17:13 |
| catherineD|2 | As current design RefStack rely on foundation member to approve (to make) a vendor to be "official" | 17:13 |
| hogepodge | RefStack shouldn't be pay to play, as a general test-result collection service it needs to have a lower barrier to entry | 17:13 |
| rockyg | Intersting. I didn't know about the sponsorship aspect. | 17:13 |
| catherineD|2 | markvoelker: ++ | 17:13 |
| hogepodge | rockyg: http://www.openstack.org/assets/marketplace/join-the-marketplace.pdf | 17:14 |
| catherineD|2 | hogepodge: RefStack will concentrating on the testing part and hope to leverage the foundation for the vendor verification parts | 17:14 |
| catherineD|2 | at least for the Mitaka cycle | 17:15 |
| hogepodge | catherineD|2: what information would you want from the marketplace for vendors? My suggestion would be to mirror that if a marketplace entry exists, otherwise create a local entry with the same fields you require | 17:15 |
| rockyg | essentially private stays private unl+ | 17:16 |
| hogepodge | catherineD|2: yeah, I can help with that. | 17:16 |
| rockyg | ess/until foundation does something | 17:16 |
| rockyg | hogepodge, catherineD|2 | 17:17 |
| catherineD|2 | I am OK with the agreement that private stays private until foundation approval | 17:17 |
| rockyg | == develop policy around it. but just leave private until then | 17:17 |
| rockyg | s/==/++ | 17:17 |
| catherineD|2 | right now we only collect name and description of vendor (and many internal meta data) | 17:18 |
| catherineD|2 | we can add additional info later if needed | 17:18 |
| rockyg | ++ | 17:18 |
| markvoelker | catherineD|2: I think maybe that's sufficent for a minimum viable product, but I might suggest one additional piece of info: the URL of the marketplace entry if one exists. That way it's super easy to provide a minimal linkage to the marketplace for situations where an entry exists. | 17:19 |
| markvoelker | (and saves you from having to actually try to pull any data from the Marketplace itself...instead you just point to it.) | 17:19 |
| catherineD|2 | markvoelker: good idea will do ... we will add a URL field as optional | 17:19 |
| eglute | does this answer your question catherineD|2? | 17:20 |
| hogepodge | catherineD|2: I want marketplace listings to link back to refstack results as part of fulfilling the transparency requirement | 17:20 |
| hogepodge | catherineD|2: that's dev effort on our side, though | 17:20 |
| catherineD|2 | it would be nice if someone can summary in an agreement sentence to show the RefStack team | 17:20 |
| markvoelker | hogepodge: Seems like if RefStack collects the marketplace link when creating a vendor it should be easy for them to provide you with the data you need to do that | 17:21 |
| markvoelker | catherineD|2: I can take that AI if no one else volunteers, but bear with me...I'm stuck in meetings the rest of the day and need to deal with the midcycle agenda too. =) | 17:22 |
| catherineD|2 | markvoelker: Thank you!! | 17:23 |
| eglute | thanks markvoelker. | 17:23 |
| markvoelker | Ok, there were several other questions in http://lists.openstack.org/pipermail/defcore-committee/2016-February/001035.html that need addressing too I believe? | 17:24 |
| eglute | catherineD|2 do you have anything else that would like resolved before midcycle? | 17:24 |
| catherineD|2 | thank you all for staying over time for RefStack discussion | 17:24 |
| catherineD|2 | eglute: I am good for now ... thank so much! | 17:24 |
| eglute | glad to help. thank you so much for doing all this work! | 17:24 |
| * markvoelker feels like he owes catherineD|2 a beer or three in Austin | 17:25 | |
| * catherineD|2 nob | 17:25 | |
| *** galstrom_zzz is now known as galstrom | 17:32 | |
| *** rockyg has quit IRC | 17:51 | |
| gema | markvoelker: ping | 17:51 |
| gema | eglute: you available? | 17:54 |
| gema | maybe I should just send an email to the list and ask all my questions | 17:54 |
| gema | :D | 17:54 |
| eglute | yes i am here! | 18:01 |
| markvoelker | gema: That'd be good--sorry, I'm in training all afternoon today | 18:01 |
| eglute | but yes, email is also good! | 18:01 |
| gema | eglute: ok, maybe you can give me a push in the right direction | 18:02 |
| eglute | i can try! | 18:02 |
| gema | https://docs.google.com/spreadsheets/d/1AEUdiv_W3QXz3R3uMPJyuZBe0cT0qE35TRAkXtvXoIs/edit#gid=0 | 18:02 |
| gema | I have gone through the keystone apis and chosen the ones I don't think require admin | 18:02 |
| gema | then looked at the tests | 18:02 |
| eglute | ok | 18:03 |
| gema | (keystone also have some tests inside the project itself) | 18:03 |
| gema | first I am not sure if the capabilities have a 1-1 relationship with the api calls | 18:03 |
| eglute | i think the tests is the tricky part | 18:04 |
| eglute | also, have you looked at https://github.com/openstack/defcore/blob/master/working_materials/scoring.txt? | 18:05 |
| eglute | and we have a script for scoring: https://github.com/openstack/defcore/tree/master/working_materials | 18:05 |
| gema | eglute: yes, however I am not sure how that ties up with the current capabilities | 18:05 |
| gema | that is just an example, right? | 18:05 |
| eglute | right | 18:06 |
| eglute | for keystone, we currently have very few tests | 18:06 |
| gema | cos the only required tests for keystone that I could find are the token ones | 18:06 |
| gema | https://github.com/openstack/defcore/blob/master/2015.07/2015.07.required.txt | 18:06 |
| eglute | https://github.com/openstack/defcore/blob/master/2016.01.json#L19 | 18:07 |
| gema | oh, so that's the right file | 18:07 |
| gema | thanks :D | 18:07 |
| gema | eglute: and regarding the name of the capability | 18:08 |
| gema | do you have a naming convention? | 18:08 |
| eglute | i dont think so... | 18:08 |
| gema | ok, will follow the identity-v2 or 3 and the name | 18:09 |
| gema | eglute: so once I have identified the calls that do not require admin and counted the tests that are calling them | 18:10 |
| gema | do we need 2 or more tests for them to be considered? | 18:10 |
| gema | wht's the minimum? | 18:10 |
| eglute | ideally, 2 or more would be good. however, right now we do have capabilities with only one test | 18:11 |
| gema | ok | 18:11 |
| gema | so I will wait for PTL's feedback and have another read at everything | 18:11 |
| gema | he said there are a few things that should be / could be included | 18:11 |
| markvoelker | Quick drive-by during a momentary break in training: | 18:12 |
| markvoelker | gema: you may want to have a look at https://github.com/openstack/defcore/blob/master/HACKING.rst and https://github.com/openstack/defcore/blob/master/working_materials/scoring.txt | 18:12 |
| gema | markvoelker: alright, will read the hacking one and the other more in depth (I have already read it but I didn't tell me a lot, now that I have tried to find the capabilities things may be different) | 18:13 |
| markvoelker | There was a fair amount of discussion about v2 vs v3 for keystone in the past (it's a bit of a trouble spot), but I don't have references handy at the moment. v2 is being deprecated now, so you can probably shy away from anything v2-related | 18:13 |
| gema | markvoelker: ok, I was expecting steve to suggest that | 18:14 |
| markvoelker | gema: cool. Thanks for getting this going! | 18:14 |
| * markvoelker returns to training | 18:15 | |
| gema | markvoelker: thank you! | 18:15 |
| gema | eglute: ok, so I will keep going and see where I end up, I was afraid I wasn't going to have anything useful for the discussion next week | 18:15 |
| gema | eglute: thank you | 18:15 |
| gema | eglute: whenever you have ready the ceilometer one , I 'd be interested in seeing it | 18:16 |
| eglute | no, thank you! i still need to do mine. i think i will have the same questions :) | 18:16 |
| gema | so that I know what format would be better | 18:16 |
| gema | :D | 18:16 |
| eglute | mine will not be a good example! i think markvoelker and hogepodge have done this more than anyone else | 18:16 |
| gema | alright, then we'll want to see theirs :D | 18:16 |
| * gema can see herself typing furiously during the 9am scoring of neutron session next Wed x) | 18:17 | |
| gema | thank you both | 18:17 |
| gema | will call it a day and continue tomorrow | 18:17 |
| * markvoelker ducks back in because who can do just one thing at a time anyway? | 18:29 | |
| markvoelker | gema: if it helps, here was the initial Neutron scoring patch...the discussion might be useful for context: https://review.openstack.org/#/c/210080/ | 18:29 |
| * markvoelker gets a stern look from the instructor and goes back to training | 18:29 | |
| *** dwalleck has joined #openstack-defcore | 18:56 | |
| *** dwalleck has quit IRC | 18:56 | |
| gema | markvoelker: it helps a lot, thanks! | 18:58 |
| *** galstrom is now known as galstrom_zzz | 18:59 | |
| *** galstrom_zzz is now known as galstrom | 19:32 | |
| *** galstrom is now known as galstrom_zzz | 19:50 | |
| *** galstrom_zzz is now known as galstrom | 20:09 | |
| *** zehicle has quit IRC | 20:15 | |
| *** zehicle has joined #openstack-defcore | 20:18 | |
| *** reed_ has joined #openstack-defcore | 20:21 | |
| *** reed_ has quit IRC | 21:35 | |
| *** edmondsw has quit IRC | 22:32 | |
| *** galstrom is now known as galstrom_zzz | 22:57 | |
| *** cjvolzka has quit IRC | 23:04 | |
| *** jmckind has joined #openstack-defcore | 23:41 | |
| *** jmckind has quit IRC | 23:42 | |
| *** jmckind has joined #openstack-defcore | 23:42 | |
| *** jmckind is now known as jmckind_ | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!